Poisson intervals for FRandR rates missed the full-record rate more than independent uniform timing predicted in 30-second circular samples from eight sleep SEEG patients

We tested nominal 95 percent Poisson intervals for FRandR rates in public sleep SEEG marks. The primary scope had 185 records from eight patients with temporal lobe epilepsy and only depth electrodes. For 30-second circular samples, the intervals included the observed full-record rate 97.56 percent of the time. A model with the same total event count and independent uniform times predicted 99.12 percent inclusion. The equal-patient excess miss fraction was 1.558 percentage points. All eight patient effects were positive. The nominal 95 percent patient bootstrap interval was 0.455 to 2.987 points. The exact two-sided sign-flip p value was 0.0078125 under independent patients and null sign symmetry. Non-circular clips gave 1.545 points. Removing the one-record patient gave 0.937 points. An exact breakpoint method avoided start-time simulation. Separate probability code reproduced the primary effect. A fresh public retrieval matched all 655 source file hashes, and a new analysis matched all ten result files byte for byte. The package preserves source tables, code, provenance, the plan, exclusions, results, and repeat instructions. This finite-rate benchmark does not measure coverage for an unknown neural intensity or test clinical outcome.

Poisson intervals · event timing · interval inclusion · high-frequency oscillations · SEEG · epilepsy · exact sampling · uniform timing model